Home > Search Results: keywords:"Mahalanobis distance."
Search:
Search Tips :: Advanced Search
Search collections:
Sort by: Display results: Output format:
National Repository of Grey Literature 2 records found  Search took 0.00 seconds. 
2D Texture Features
 Pasáček, Václav ; Šiler, Ondřej (referee) ; Švub, Miroslav (advisor)
Because texture of object is very valuable information in computer vision, it is important to describe it somehow. And for this serve texture features. Optimal selection of features is very important for recognizing texture. In this bachelor thesis were used local binary patterns (LBP) as a method of gaining texture feature. In this method is not its value the texture feature, but histogram of percent occurrence values in the entire texture. To compare histograms there is used Euclidean distance, Bhattacharyya distance or Mahalanobis distance. Main purpose of this thesis is mutually comparing of texture clasification by several variants of LBP and evaluation of their outcomes by Euclidean distance, Bhattacharyya distance or Mahalanobis distance.
Detailed record
2D Texture Features
 Pasáček, Václav ; Šiler, Ondřej (referee) ; Švub, Miroslav (advisor)
Because texture of object is very valuable information in computer vision, it is important to describe it somehow. And for this serve texture features. Optimal selection of features is very important for recognizing texture. In this bachelor thesis were used local binary patterns (LBP) as a method of gaining texture feature. In this method is not its value the texture feature, but histogram of percent occurrence values in the entire texture. To compare histograms there is used Euclidean distance, Bhattacharyya distance or Mahalanobis distance. Main purpose of this thesis is mutually comparing of texture clasification by several variants of LBP and evaluation of their outcomes by Euclidean distance, Bhattacharyya distance or Mahalanobis distance.
Detailed record

Interested in being notified about new results for this query?
Subscribe to the RSS feed.
Digital Repository :: Search :: Submit :: Personalize :: Help
Powered by Invenio v1.1.2
Maintained by admin.invenio@techlib.cz
This site is also available in the following languages:
Česky  English